Inter-vehicle network false risk warning identification method and vehicle-mounted electronic device using same

Abstract

The invention discloses an inter-vehicle network false risk warning identification method and a vehicle-mounted electronic device using the same, and the inter-vehicle network false risk warning identification method comprises the steps: obtaining a first risk warning based on an inter-vehicle network in response to obtaining a first risk warning based on the inter-vehicle network; judging whether a second risk warning based on object detection operation corresponding to the first risk warning is obtained or not; in response to determining that the second risk warning is obtained, determining that the first risk warning is trusted; in response to determining that the second risk warning is not obtained, determining whether the first risk warning is trusted according to the object detection operation; in response to determining that the first risk alert is trusted, adjusting a driving behavior of the vehicle in accordance with the first risk alert or the second risk alert; and ignoring the first risk alert in response to determining that the first risk alert is untrustworthy.

Description

Technical Field

[0001] The present invention relates to a method for identifying false risk warnings, and particularly to a method for identifying false risk warnings in a workshop network and an in-vehicle electronic device using the method. Background Art

[0002] Vehicular Ad-hoc Network (VANET) can enable Vehicle-to-Everything (V2X) communication and provide support for intelligent transportation systems. Therefore, the importance of the network security of VANET has been increasing. There are many network attacks that are carried out via V2X. Among them, the most widespread and serious network attack is the Sybil attack.

[0003] The Sybil attack can create false Vehicle-to-Vehicle (V2V) information that can be propagated via V2X by hackers locating and modifying vehicle-related information. For example, fabricating and spreading a set of non-existent false workshop network data to attempt to influence the driving behavior of vehicles that receive the false workshop network data and thus generate risk warning information, thereby reducing the driving safety of the vehicles and resulting in losses of life and property. Summary of the Invention

[0004] An embodiment of the present invention provides a method for identifying false risk warnings in a workshop network, which is applicable to an in-vehicle electronic device of a vehicle. The in-vehicle electronic device includes a processor and a communication circuit unit, wherein the in-vehicle electronic device is connected to the workshop network via the communication circuit unit. The method includes: instructing an advanced driver assistance system of the vehicle to perform object detection operation; in response to obtaining a first risk warning based on the workshop network, determining whether a second risk warning based on the object detection operation corresponding to the first risk warning is obtained, wherein the second risk warning is received from the advanced driver assistance system; in response to determining that the second risk warning based on the object detection operation corresponding to the first risk warning is obtained, determining that the first risk warning based on the workshop network is trustworthy; in response to determining that the second risk warning based on the object detection operation corresponding to the first risk warning is not obtained, determining whether the first risk warning based on the workshop network is trustworthy according to the object detection result of the object detection operation and a plurality of preset conditions corresponding to the object detection operation; in response to determining that the first risk warning based on the workshop network is trustworthy, adjusting the driving behavior of the vehicle according to the first risk warning or the second risk warning; and in response to determining that the first risk warning based on the workshop network is not trustworthy, ignoring the first risk warning and not adjusting the driving behavior of the vehicle according to the first risk warning.

[0005] Another embodiment of the present invention provides a vehicle-mounted electronic device applicable to a vehicle, including: a communication circuit unit for connecting to a vehicle-to-vehicle network; a processor coupled to the communication circuit unit; and a storage circuit unit storing instructions, wherein the instructions are configured, when executed by the processor, to: instruct an advanced driver assistance system of the vehicle to perform object detection operations; in response to obtaining a first risk warning based on the vehicle-to-vehicle network, determine whether a second risk warning based on the object detection operations corresponding to the first risk warning is obtained, wherein the second risk warning is received from the advanced driver assistance system; in response to determining that the second risk warning based on the object detection operations corresponding to the first risk warning is obtained, determine that the first risk warning based on the vehicle-to-vehicle network is trustworthy; in response to determining that the second risk warning based on the object detection operations corresponding to the first risk warning is not obtained, determine whether the first risk warning based on the vehicle-to-vehicle network is trustworthy according to the object detection results of the object detection operations and a plurality of preset conditions corresponding to the object detection operations; in response to determining that the first risk warning based on the vehicle-to-vehicle network is trustworthy, adjust the driving behavior of the vehicle according to the first risk warning or the second risk warning; and in response to determining that the first risk warning based on the vehicle-to-vehicle network is not trustworthy, ignore the first risk warning and do not adjust the driving behavior of the vehicle according to the first risk warning.

[0006] Based on the above, the vehicle-to-vehicle network false risk warning identification method provided by the present invention and the vehicle-mounted electronic device using the method can determine whether the first risk warning is trustworthy according to the object detection operations, the first risk warning based on the vehicle-to-vehicle network, and the second risk warning based on the object detection operations, and then identify the false first risk warning, avoiding wrong driving behaviors corresponding to the false first risk warning and ensuring the safety of the vehicle. In addition, since the first risk warning is only verified for trustworthiness when it is obtained, it is not necessary to continuously compare objects in all surrounding environments at all times to avoid risk warnings caused by false vehicle data, reducing the resource consumption of the vehicle-mounted electronic device and thus improving the driving efficiency. Description of the Drawings

[0007] Figure 1 is a block diagram of a plurality of electronic devices configured on a vehicle shown in an embodiment of the present invention;

[0008] Figure 2 is an operation flowchart of a vehicle-to-vehicle network false risk warning identification method shown in an embodiment of the present invention;

[0009] Figure 3 is another operation flowchart of a vehicle-to-vehicle network false risk warning identification method shown in an embodiment of the present invention;

[0010] Figure 4A is the flowchart of step S350 illustrated in an embodiment of the present invention; Figure 3 of the flowchart of step S350;

[0011] Figure 4B is the flowchart of step S360 illustrated in an embodiment of the present invention; Figure 3 of the flowchart of step S360;

[0012] Figure 4C is the flowchart of step S370 illustrated in an embodiment of the present invention; Figure 3 of the flowchart of step S370;

[0013] Figure 5A is a schematic diagram of a plurality of electronic devices connected to a workshop network illustrated in an embodiment of the present invention;

[0014] Figure 5B is a schematic diagram of the images and object detection results obtained by an in-vehicle electronic device illustrated in an embodiment of the present invention.

[0015] Symbol Description

[0016] 10: Vehicle

[0017] 100: In-vehicle electronic device

[0018] 200: Advanced driver assistance system

[0019] 300: Driving system

[0020] 400: Workshop network

[0021] 110: Processor

[0022] 120: Communication circuit unit

[0023] 130: Connection interface

[0024] 140: Storage circuit unit

[0025] WD1, WD2: Risk warning

[0026] OD: Object detection result

[0027] S210, S220, S230, S240, S250, S260, S270: Process steps of the method for identifying false risk warnings in a workshop network

[0028] S310, S320, S330, S340, S350, S360, S370, S380, S390, S400: Another process steps of the method for identifying false risk warnings in a workshop network

[0029] S351, S352, S353:Figure 3 The process steps of step S350

[0030] S361, S362, S363: Figure 3 The process steps of step S360

[0031] S371, S372, S373, S374: Figure 3 The process steps of step S370

[0032] TA: Certification Authority

[0033] RSU: Road Side Unit

[0034] C1, C2, C3: Vehicle / Object

[0035] FCW: Forward Collision Warning

[0036] BSW: Blind Spot Warning

[0037] IMG1, IMG2: Image

[0038] FOV: Field of View

[0039] BB: Bounding Box Detailed implementation manners

[0040] Please refer to Figure 1 , in this embodiment, the vehicle 10 includes: an in-vehicle electronic device 100, an advanced driver assistance system 200, and a driving system 300. The in-vehicle electronic device 100 includes a processor 110, a communication circuit unit 120, a connection interface 130, and a storage circuit unit 140. The processor 110 is coupled to the communication circuit unit 120, the connection interface 130, and the storage circuit unit 140. The in-vehicle electronic device 100 can also be referred to as an On Board Unit (OBU).

[0041] The advanced driver assistance system 200 includes a plurality of sensors (such as cameras, radars, lidars, ultrasonic transceivers, GPS receivers, accelerometers, inertial meters, gyroscopes, etc.) and a logic operation unit (such as a processor or an MCU) to plan / assist the driving behavior of the vehicle 10 and provide corresponding information to the in-vehicle electronic device 100. The driving system 300 is used to control the movement of the vehicle 10. The advanced driver assistance system (Advanced Driver Assistance Systems, ADAS) 200 and the driving system 300 can also be integrated into the in-vehicle electronic device 100 so that the in-vehicle electronic device 100 can control all operations of the vehicle 10.

[0042] The processor 110 is, for example, a Microprogrammed Control Unit (MCU), a Central Processing Unit (CPU), a programmable microprocessor, an Application Specific Integrated Circuits (ASIC), a Programmable Logic Device (PLD), or other similar devices.

[0043] The communication circuit unit 120 is coupled to the processor 110 for transmitting or receiving data by wireless communication. In this embodiment, the communication circuit unit 120 may have a wireless communication circuit module (not shown) and support one or a combination of a Global System for Mobile Communication (GSM) system, a Wireless Fidelity (WiFi) system, different generations of mobile communication technologies (such as 3G - 6G), and Bluetooth communication technology, and is not limited thereto. The communication circuit unit 120 is used to connect to the vehicle-to-vehicle network 400 via the Vehicle-to-Vehicle (V2V) and Vehicle-to-everything (V2X) communication protocols. The vehicle-to-vehicle network 400 is, for example, a Vehicular ad-hoc Network (VANET). The processor 110 may receive surrounding vehicle-related data (such as vehicle network data VD such as speed, position, driving direction, braking, and loss of stability, as shown in Figure 1 ), via the vehicle-to-vehicle network 400, and execute a preset application program to obtain / generate a risk warning message WD1 (also referred to as the first risk warning WD1), that is, obtain the first risk warning WD1 based on the vehicle-to-vehicle network 400. In one embodiment, the processor 110 may generate the first risk warning WD1 based on the received vehicle network data VD. It should be noted that, in this embodiment, the processor 110 generates the first risk warning WD1 based on the vehicle network data VD from the vehicle-to-vehicle network 400, but the present invention is not limited thereto. For example, in another embodiment, the processor 110 may also receive the first risk warning WD1 from the vehicle-to-vehicle network 400.

[0044] When it is determined that the first risk warning WD1 based on the vehicle workshop network 400 is trustworthy, the processor 110 can generate and transmit a corresponding control instruction CS to the driving system 300 to adjust the driving behavior of the vehicle 10. In addition, in another embodiment, when it is determined that the first risk warning WD1 based on the vehicle workshop network 400 is trustworthy, the processor 110 notifies the advanced driver assistance system 200, so that the advanced driver assistance system 200 generates and transmits a corresponding control instruction CS to the driving system 300 to adjust the driving behavior of the vehicle 10.

[0045] The connection interface 130 is coupled to the processor 110. The processor 110 is used to establish a data connection with the advanced driver assistance system 200 via the connection interface 130 to transmit data to and from the advanced driver assistance system 200. For example, receiving the object detection result OD and the risk warning information WD2 (also referred to as the second risk warning) from the advanced driver assistance system 200. In one embodiment, the connection interface 130 includes an in-vehicle Ethernet.

[0046] In this embodiment, the types of the first risk warning WD1 and the second risk warning WD2 include: Forward Collision Warning (FCW); and Blind Spot Warning (BSW). However, the present invention is not limited thereto. For example, the types of the first risk warning WD1 and the second risk warning WD2 may further include other types of risk warnings that can be determined by the processor 100 or the advanced driving system 200 according to the object detection operation or the sensed data collected.

[0047] The storage circuit unit 140 is coupled to the processor 110. The storage circuit unit 140 can store data under the instruction of the processor 110. The data includes external data, such as vehicle workshop network data; and internal data and system data. The system data is, for example, software / firmware for processing vehicle workshop network data from the vehicle workshop network, software / firmware for processing the object detection data of the vehicle itself, etc. The internal data is, for example, object detection data, etc. The storage circuit unit includes any type of hard disk drive (HDD) or non-volatile memory storage device (such as SSD). In one embodiment, the storage circuit unit further includes a memory for temporarily storing the instructions or data executed by the processor, such as Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), etc.

[0048] In one embodiment, the in-vehicle electronic device 100 further includes an input / output unit, which includes an input device and an output device. The input device is, for example, a microphone, a touchpad, a touch panel, a knob, a button, or other devices, which are used to allow a user to input data or control functions that the user desires to operate. The output device is, for example, a display, a speaker, or other devices, and the present case is not limited thereto. In one embodiment, the input / output unit may be a touch screen, a head-up display, or a head-mounted display.

[0049] Please refer to Figure 2 , in step S210, the processor 110 instructs the advanced driver assistance system 200 of the vehicle 10 to perform an object detection operation. In one embodiment, in response to determining that the advanced driver assistance system 200 does not perform an object detection operation, the processor 110 instructs the advanced driver assistance system 200 to perform an object detection operation.

[0050] Next, in step S220, in response to obtaining a first risk warning WD1 based on the vehicle-to-vehicle network 400, the processor 110 determines whether a second risk warning WD2 based on the object detection operation corresponding to the first risk warning WD1 is obtained, where the second risk warning WD2 is received from the advanced driver assistance system 200. Briefly speaking, when vehicle-to-vehicle network data VD is received from the vehicle-to-vehicle network 400 and a corresponding first risk operation WD1 is generated (i.e., a first risk warning WD1 based on the vehicle-to-vehicle network 400 is obtained), the processor 110 will correspondingly determine whether a corresponding second risk warning WD2 is also received / obtained from the advanced driver assistance system 200. In one embodiment, the time difference between obtaining the second risk warning WD2 and obtaining the first risk warning WD1 needs to be less than a preset time threshold value (e.g., 5 seconds or other seconds).

[0051] Next, in step S230, in response to determining that the second risk warning WD2 based on the object detection operation corresponding to the first risk warning WD1 is obtained, the processor 110 determines that the first risk warning WD1 based on the vehicle-to-vehicle network 400 is trustworthy. That is to say, when the first risk warning WD1 is obtained, if the processor 110 also determines that the second risk warning WD2 corresponding to the first risk warning WD1 is obtained from the advanced driver assistance system 200 (for example, both the first risk warning WD1 and the second risk warning WD2 indicate the same type or the same direction of risk around the vehicle 10), the processor 110 trusts this first risk warning WD1. In other words, since the second risk warning WD2 transmitted by the advanced driver assistance system 200 is a judgment based on the collected sensing data, it is more difficult to be forged and has a higher credibility. The processor 110 uses the second risk warning WD2 to verify the first risk warning WD1 (because the first risk warning WD1 may be a risk warning generated by false information transmitted to the in-vehicle electronic device 100 via the vehicle-to-vehicle network 400).

[0052] More specifically, the steps of determining whether a second risk warning WD2 based on object detection operation corresponding to the first risk warning WD1 is obtained include: determining whether the type of the first risk warning WD1 is the same as the type of the received second risk warning WD2; in response to determining that the type of the first risk warning WD1 is not the same as the type of the received second risk warning WD2, determining that the second risk warning WD2 based on object detection operation corresponding to the first risk warning WD1 is not obtained; in response to determining that the type of the first risk warning WD1 is the same as the type of the received second risk warning WD2, determining whether the risk object of the first risk warning WD1 corresponds to the risk object of the second risk warning WD2; in response to determining that the risk object of the first risk warning WD1 corresponds to the risk object of the second risk warning WD2, determining that the second risk warning WD2 based on object detection operation corresponding to the first risk warning WD1 is obtained; and in response to determining that the risk object of the first risk warning WD1 does not correspond to the risk object of the second risk warning WD2, determining that the second risk warning WD2 based on object detection operation corresponding to the first risk warning WD1 is not obtained.

[0053] Briefly, in response to obtaining the second risk warning WD2 before and after the time point of obtaining the first risk warning WD1. The processor 110 may first determine whether the first risk warning WD1 and the second risk warning WD2 belong to the same type (or the same relative position). If so, it may further determine whether the object of the first risk warning WD1 is similar to or equivalent to the object of the second risk warning WD2. If so, the processor 110 determines that the obtained second risk warning WD2 corresponds to the first risk warning WD1. If one of the above judgments results in a negative result, the processor may determine that the second risk warning WD2 corresponding to the first risk warning WD1 is not obtained.

[0054] Next, in step S240, in response to determining that the second risk warning WD2 based on object detection operation corresponding to the first risk warning WD1 is not obtained, it is determined whether the first risk warning WD1 based on the workshop network 400 is trustworthy according to the object detection result of the object detection operation and a plurality of preset conditions corresponding to the object detection operation. In this embodiment, the object detection operation includes one or more of the following operations: a first object detection operation based on image data; a second object detection operation based on point cloud data; and a third object detection operation based on mixed data. The mixed data includes image data and point cloud data. In another embodiment, the mixed data is fusion data generated from image data and point cloud data.

[0055] The first object detection operation based on image data (or pixel array) is more proficient in determining the area of the first object but not in determining the relative distance between the first object and the vehicle. In contrast, the second object detection operation based on point cloud data is less proficient in determining the area of the second object but more proficient in determining the relative distance between the second object and the vehicle. On the other hand, the third object detection operation based on mixed data has good reliability in both the area determination and the relative distance determination of the third object, but it requires additional sensing data and computing resources to process the fusion operation of image data and point cloud data.

[0056] More specifically, the processor 110 can first identify the type of object detection operation and, according to different types of object detection operations, use different preset conditions to analyze the corresponding object detection results to determine whether the first risk warning WD1 based on the vehicle-to-vehicle network 400 is trustworthy. In other words, since the advanced driver assistance system 200 makes object detection results based on the collected sensing data, which is difficult to forge and has higher credibility, the processor 110 can also use the received object detection results and the corresponding preset conditions for analysis to determine whether the received object detection results have an object corresponding to the first risk warning WD1, thereby verifying the first risk warning WD1 (because the first risk warning WD1 may be a risk warning generated by false information transmitted to the in-vehicle electronic device 100 via the vehicle-to-vehicle network 400).

[0057] Next, in step S250, in response to determining that the first risk warning WD1 based on the vehicle-to-vehicle network 400 is trustworthy, the processor 110 adjusts the driving behavior of the vehicle according to the first risk warning WD1 or the second risk warning WD2.

[0058] Specifically, in one embodiment, after determining that the first risk warning information WD1 is trustworthy based on the second risk warning information WD2 corresponding to the first risk warning information WD1, the processor 110 notifies the advanced driver assistance system 200 that the first risk warning WD1 is trustworthy, and the advanced driver assistance system 200 generates a corresponding control instruction CS based on the second risk warning WD2 to send the generated control instruction CS to the driving system 300 to adjust the driving behavior of the vehicle. For example, assuming that the first risk warning WD1 and the second risk warning WD2 belong to a forward collision warning, the generated control instruction CS is used to instruct the driving system 300 to perform a braking action, reduce the driving speed, or change lanes, so as to avoid the risk events corresponding to the first risk warning WD1 and the second risk warning WD2 by adjusting the driving behavior of the vehicle 10.

[0059] In another embodiment, after determining that the first risk warning message WD1 is trustworthy, the processor 110 may also generate a corresponding control instruction CS based on the first risk warning WD1 to send the generated control instruction CS to the driving system 300 to adjust the driving behavior of the vehicle. For example, assuming that the trustworthy first risk warning WD1 belongs to a forward collision warning, the control instruction CS is used to instruct the driving system 300 to perform a braking action, reduce the driving speed, or change lanes, so as to avoid the risk event corresponding to the first risk warning WD1 by adjusting the driving behavior of the vehicle 10.

[0060] In contrast, in step S260, in response to determining that the first risk warning WD1 based on the vehicle-to-vehicle network 400 is untrustworthy, the processor 110 ignores the first risk warning WD1 and does not adjust the driving behavior of the vehicle 10 according to the first risk warning WD1. Then, in step S270, the processor 110 reports the first risk warning WD1 to the certification authority (such as Figure 5A the certification authority TA shown) via the vehicle-to-vehicle network 400. For example, in one embodiment, the processor 110 may report the vehicle-to-vehicle data that caused the generation of the untrustworthy risk warning and / or the untrustworthy risk warning to the certification authority TA so that the certification authority TA can record it and further monitor or manage these data and their sources.

[0061] The following uses Figure 3 and Figures 4A to 4C to elaborate on other process details.

[0062] Please refer to Figure 3 , the advanced driver assistance system 200 performs object detection operations (S310). Then, the processor 110 determines whether a first risk warning WD1 based on the vehicle-to-vehicle network 400 is obtained (S320). If so, it then determines whether a second risk warning WD2 based on the object detection operation corresponding to the first risk warning WD1 is obtained (S330). If so, the processor 110 adjusts the driving behavior of the vehicle according to the first risk warning or the second risk warning (S380); if not, the processor 110 further identifies the type of the object detection operation (S340). The processor 110 can obtain and identify the results and related information of the performed object detection operation from the advanced driver assistance system 200 (such as, object detection results, types of object detection operations).

[0063] According to different types of object detection operations, the processor 110 will adopt different analysis and judgment methods to determine whether the first risk warning WD1 is trustworthy. Specifically, when the object detection operation is the first object detection operation based on image data, the processor 110 executes step S350; when the object detection operation is the first object detection operation based on image data, the processor 110 executes step S360; when the object detection operation is the first object detection operation based on image data, the processor 110 executes step S370.

[0064] More specifically, please refer to Figure 4A , when the object detection operation is the first object detection operation based on image data, the steps of determining whether the first risk warning WD1 based on the workshop network 400 is trustworthy according to the object detection result of the object detection operation and multiple preset conditions corresponding to the object detection operation include steps S351 to S353.

[0065] In step S351, the processor 110 identifies the first object detection result of the first object detection operation, as well as the first object feature threshold value and the first object probability threshold value corresponding to the first object detection operation. Among them, the first object detection result includes the detected first object, the first object feature, and the first object probability corresponding to the detected first object, and the first object feature threshold value and the first object probability threshold value are multiple first preset conditions corresponding to the first object detection operation.

[0066] In this embodiment, when the advanced driver assistance system 200 initiates the second risk warning WD2 by executing the first object detection operation, the processor 110 can record the object feature value and the object probability value of the target object corresponding to the second risk warning WD2 at that time. Then, using the recorded historical data, the first object feature threshold value and the first object probability threshold value are trained through a machine learning algorithm, or the first object feature threshold value and the first object probability threshold value are calculated through statistics (such as calculating the average value or median). The obtained first object feature threshold value and the first object probability threshold value can be set as multiple first preset conditions corresponding to the first object detection operation. In addition, in one embodiment, different vehicle speeds can be matched to different first object feature threshold values and first object probability threshold values.

[0067] Please go back to Figure 4A , then, in step S352, the processor 110 determines whether the first object feature is greater than the first object feature threshold value.

[0068] Then, in step S353, the processor 110 determines whether the first object probability is greater than the first object probability threshold value.

[0069] In response to determining that the first object feature is greater than the first object feature threshold (S352 determines "yes") and the first object probability is greater than the first object probability threshold (S353 determines "yes"), the processor 110 determines that the first risk warning WD1 based on the vehicle network 400 is trustworthy.

[0070] In contrast, in response to determining that the first object feature is not greater than the first object feature threshold (S352 determines "no") or the first object probability is not greater than the first object probability threshold (S353 determines "no"), the processor 110 determines that the first risk warning WD1 based on the vehicle network 400 is untrustworthy.

[0071] It should be noted that in another embodiment, steps S352 and S353 can be swapped in order (i.e., first execute step S353, and then execute step S352). In addition, in this embodiment, the first object feature includes the size or ratio of the image features of the first detected object, where the ratio of the image features of the first detected object is the ratio of the size of the image features of the first detected object to the preset image size of the image data, and the first object probability is the probability of the existence of the first detected object.

[0072] The size of the image feature is, for example, the size of the image area covered by the bounding box marking the first detected object. The area size can be compared with the area size of the image captured by the advanced driver assistance system 200 for the environment of the vehicle 10 (i.e., the preset image size) to obtain a ratio, that is, the ratio of the image features.

[0073] On the other hand, please refer to Figure 4B , when the object detection operation is the second object detection operation based on point cloud data, the steps of determining whether the first risk warning WD1 based on the vehicle network 400 is trustworthy according to the object detection result of the object detection operation and multiple preset conditions corresponding to the object detection operation include steps S361 to S363.

[0074] In step S361, the processor 110 identifies the second object detection result of the second object detection operation, as well as the second object distance threshold and the second object probability threshold corresponding to the second object detection operation. Among them, the second object detection result includes the detected second object, the second object distance, and the second object probability corresponding to the detected second object, and the second object distance threshold and the second object probability threshold are multiple second preset conditions corresponding to the second object detection operation.

[0075] In this embodiment, when the advanced driver assistance system 200 initiates the second risk warning WD2 via performing the second object detection operation, the processor 110 can record the object distance value and the object probability value of the target object corresponding to the second risk warning WD2 at that time. Then, using the recorded historical data, the second object distance threshold value and the second object probability threshold value are trained through a machine learning algorithm, or the second object distance threshold value and the second object probability threshold value are calculated through statistics.

[0076] Next, in step S362, the processor 110 determines whether the second object distance is less than the second object distance threshold value.

[0077] Next, in step S363, the processor 110 determines whether the second object probability is greater than the second object probability threshold value.

[0078] In response to determining that the second object distance is less than the second object distance threshold value (S362 determines "yes") and the second object probability is greater than the second object probability threshold value (S363 determines "yes"), the processor 110 determines that the first risk warning WD1 based on the vehicle-to-vehicle network 400 is trustworthy.

[0079] In response to determining that the second object distance is not less than the second object distance threshold value (S362 determines "no") or the second object probability is not greater than the second object probability threshold value (S363 determines "no"), the processor 110 determines that the first risk warning WD1 based on the vehicle-to-vehicle network 400 is untrustworthy.

[0080] It should be noted that, in another embodiment, the order of steps S362 and S363 can be swapped. In addition, in this embodiment, the second object distance is the distance between the detected second object and the vehicle 10, where the second object probability is the probability of the existence of the detected second object.

[0081] On the other hand, please refer to Figure 4C , when the object detection operation is the third object detection operation based on mixed data, the steps of determining whether the first risk warning WD1 based on the vehicle-to-vehicle network 400 is trustworthy according to the object detection result of the object detection operation and multiple preset conditions corresponding to the object detection operation include steps S371 to S374.

[0082] In step S371, the processor 110 identifies the third object detection result of the third object detection operation and the third object feature threshold, the third object distance threshold, and the third object probability threshold corresponding to the third object detection operation. Among them, the third object detection result includes the detected third object and the third object feature, the third object distance, and the third object probability corresponding to the detected third object. The third object feature threshold, the third object distance threshold, and the third object probability threshold are multiple third preset conditions corresponding to the third object detection operation. Similar to the obtaining methods of the above-mentioned first object feature threshold, the first object probability threshold, the second object distance threshold, and the second object probability threshold, the third object feature threshold, the third object distance threshold, and the third object probability threshold can be obtained through training by a machine learning algorithm or calculated through statistics.

[0083] Next, in step S372, the processor 110 determines whether the third object probability is greater than the third object probability threshold. In response to determining that the third object probability is greater than the third object probability threshold (S372 determination is "yes"), the processor 110 executes step S373; in response to determining that the third object probability is not greater than the third object probability threshold (S372 determination is "no"), the processor 110 determines that the first risk warning WD1 based on the in-vehicle network 400 is untrustworthy.

[0084] Next, in step S373, the processor 110 determines whether the third object feature is greater than the third object feature threshold. In response to determining that the third object feature is greater than the third object feature threshold (S373 determination is "yes"), the processor 110 determines that the first risk warning WD1 based on the in-vehicle network 400 is trustworthy; in response to determining that the third object feature is not greater than the third object feature threshold (S373 determination is "no"), the processor 110 executes step S374.

[0085] Next, in step S374, the processor 110 determines whether the third object distance is less than the third object distance threshold. In response to determining that the third object distance is less than the third object distance threshold (S374 determination is "yes"), the processor 110 determines that the first risk warning WD1 based on the in-vehicle network 400 is trustworthy; in response to determining that the third object distance is not less than the third object distance threshold (S374 determination is "no"), the processor 110 determines that the first risk warning WD1 based on the in-vehicle network 400 is untrustworthy.

[0086] It should be noted that, in another embodiment, steps S373 and S374 can be swapped with each other. In addition, in this embodiment, the third object feature includes the detected image feature size or image feature ratio of the third object, where the detected image feature ratio of the third object is the ratio of the detected image feature size of the third object to the preset image size of the image data, and the third object probability is the probability of the existence of the detected third object.

[0087] In this embodiment, the detected first object, the detected second object, and the detected third object can also be referred to as the target first object, the target second object, and the target third object, which are, for example, the objects most likely to cause a risk warning. For example, the vehicle in front or the vehicle beside in the blind spot of the perspective. In one embodiment, the detected first object, the detected second object, and the detected third object can also be the objects closest to the vehicle itself.

[0088] Please return to Figure 3 , in response to determining that the first risk warning WD1 based on the vehicle-to-vehicle network 400 is trustworthy, the processor 110 adjusts the driving behavior of the vehicle according to the first risk warning or the second risk warning (S380); in response to determining that the first risk warning WD1 based on the vehicle-to-vehicle network 400 is untrustworthy, the processor 110 ignores the first risk warning WD1 and does not adjust the driving behavior of the vehicle 10 according to the first risk warning WD1 (S390). In addition, the processor 110 can report the first risk warning WD1 (or the false vehicle-to-vehicle data corresponding to the first risk warning WD1) to the certification authority (S400).

[0089] For example, please refer to Figure 5A and Figure 5B , the vehicle-to-vehicle network (such as, VANET) includes three main entities, namely, the on-board unit OBU of the vehicle, the certification authority TA, and the roadside unit RSU. Assume that the OBUs (i.e., in-vehicle electronic devices) of vehicles C1 to C3, the certification authority TA, and the roadside unit RSU each have a wireless communication circuit unit to connect to the vehicle-to-vehicle network. The certification authority TA is responsible for authorizing the roadside units RSU and OBUs on the road. The communication between the OBU and the RSU includes vehicle-to-vehicle communication (V2V) and vehicle-to-infrastructure communication (V2I). A spoofing attack can send false vehicle-to-vehicle network data VD to the OBU in the vehicle-to-vehicle network through the security application of V2V, so that the OBU generates a wrong risk warning.

[0090] In addition, assume that vehicle C1 is the host vehicle 10, and the object detection operation performed by its advanced driver assistance system 200 is the first object detection operation based on image data. The camera of the advanced driver assistance system 200 captures the image IMG1 via the field of view FOV. The advanced driver assistance system 200 can perform the first object detection operation on the image IMG1, identify the target first object C2, and use the bounding box BB to mark the target first object C2. In addition, the advanced driver assistance system 200 can determine the image feature size or image feature ratio corresponding to the target first object C2 through the bounding box BB or the image of the target first object C2. The image feature ratio of the target first object C2 is, for example, the ratio obtained by dividing the size of the image IMG2 covered by the bounding box BB by the size of the image IMG1. Information such as the target first object C2, the probability of the existence of the corresponding target first object C2 (i.e., the target first object probability), and the image feature size or image feature ratio corresponding to the target first object C2 (i.e., the target first object feature) can be packed into the object detection result OD and transmitted to the in-vehicle electronic device 100 of the vehicle C1.

[0091] In the first example, assume that vehicle C1 has performed the first object detection operation and obtained a forward collision warning (the first risk warning WD1) based on the vehicle-to-vehicle network 400, which indicates that it is about to hit the vehicle C2 (FCW) in front. At the same time, the advanced driver assistance system 200 of vehicle C1 sends the object detection result OD corresponding to vehicle C2 and believes that vehicle C2 does not meet the activation condition of the forward collision warning of the advanced driver assistance system 200 (no second risk warning WD2 corresponding to vehicle C2 is sent). In addition, the advanced driver assistance system 200 of vehicle C1 sends the object detection result OD corresponding to vehicle C3 and believes that vehicle C3 meets the activation condition of the blind spot warning of the advanced driver assistance system 200. The advanced driver assistance system 200 senses the risk of vehicle C3 located in the blind spot area via the detection range DR of the BSW and sends a blind spot warning (the second risk warning WD2 corresponding to vehicle C3) to the in-vehicle electronic device 100 of vehicle C1. That is, the second risk warning sent by the advanced driver assistance system 200 of vehicle C1 does not correspond to the first risk warning (because the second risk warning is for the BSW of vehicle C3 and does not correspond to the FCW of vehicle C2 warned by the first risk warning).

[0092] In this case (when the determination in S330 is "no" because the obtained second risk warning WD2 does not correspond to the first risk warning WD1), vehicle C1 will further identify whether the received object detection result OD belongs to the first object detection operation based on image data (S340). At this time, vehicle C1 can access multiple first preset conditions corresponding to the first object detection operation, which include the first object feature threshold value and the first object probability threshold value.

[0093] Next, vehicle C1 further determines whether the first risk warning corresponding to vehicle C2 is trustworthy (S350) based on the first object detection result of the first object detection operation based on image data (such as, the image feature ratio of the target first object C2 and the existence probability of the target first object C2) and the corresponding first preset conditions (such as, the first object feature threshold value and the first object probability threshold value).

[0094] When the image feature ratio of the target first object C2 is not greater than the first object feature threshold value or the existence probability of the target first object C2 is not greater than the first object probability threshold value, the processor 110 determines that the first risk warning WD1 is not trustworthy, that is, the first risk warning WD1 may be forged. The processor 110 will then ignore this first risk warning WD1 (S390). In addition, the processor 110 further reports the first risk warning WD1 to the certification authority (S400). For example, in one embodiment, the processor 110 will further generate a false risk warning report and transmit the false risk warning report (such as, including false in-vehicle network data and / or the corresponding generated false risk warning) to the certification authority TA via the communication circuit unit 120 (S400), so that the certification authority TA can record the current false risk warning and / or the corresponding false in-vehicle network data VD in the in-vehicle network 400, and then the certification authority TA can decide whether to blacklist or block this false risk warning / or the corresponding false in-vehicle network data VD (and its corresponding entity / source). In this way, the security of the overall in-vehicle network can be better maintained.

[0095] Conversely, when the image feature ratio of the target first object C2 is greater than the first object feature threshold value and the existence probability of the target first object C2 is greater than the first object probability threshold value, the processor 110 determines that the first risk warning WD1 is trustworthy, that is, the first risk warning WD1 is not forged. The processor 110 will then adjust the driving behavior of vehicle C1 according to the first risk warning WD1 (S380). For example, generate and send a corresponding control instruction CS to the driving system 300 according to the first risk warning WD1 to adjust the driving behavior of vehicle C1.

[0096] In a second example, assume that vehicle C1 has performed a first object detection operation and obtained a forward collision warning (first risk warning WD1) based on the vehicle-to-vehicle network 400, which indicates that it is about to hit the vehicle C2 ahead (FCW). On the other hand, further assume that at the same time, the advanced driver assistance system 200 of vehicle C1 sends the object detection result OD corresponding to vehicle C2 and determines that vehicle C2 meets the activation condition for the forward collision warning of the advanced driver assistance system 200. The advanced driver assistance system 200 sends a forward collision warning (second risk warning) corresponding to vehicle C2 (FCW) to the in-vehicle electronic device 100 of vehicle C1.

[0097] In this case (S330 determines "yes" because the obtained second risk warning WD2 corresponds to the first risk warning WD1), the processor 110 of vehicle C1 determines that the first risk warning WD1 corresponding to vehicle C2 is trustworthy. Then, the processor 110 adjusts the driving behavior of vehicle C1 according to the first risk warning WD1 (S380). For example, it notifies the advanced driver assistance system 200 that the first risk warning WD1 is trustworthy, so that the advanced driver assistance system 200 can generate and send a control instruction CS to the driving system 300 based on the second risk warning WD2 to adjust the driving behavior of vehicle C1.

[0098] In a third example, assume that vehicle C1 has performed a first object detection operation and obtained a forward collision warning (first risk warning WD1) based on the vehicle-to-vehicle network 400, which indicates that it is about to hit the vehicle C2 ahead (FCW). On the other hand, further assume that at the same time, the advanced driver assistance system 200 of vehicle C1 sends the object detection result OD corresponding to vehicle C2 and determines that vehicle C2 does not meet the activation condition for the forward collision warning of the advanced driver assistance system 200. The advanced driver assistance system 200 does not send a forward collision warning (second risk warning) corresponding to vehicle C2 (FCW) to the in-vehicle electronic device 100 of vehicle C1. That is, the advanced driver assistance system 200 of vehicle C1 does not send a second risk warning corresponding to the first risk warning to the processor 110 of vehicle C1.

[0099] In this case (S330 determines "no" because no second risk warning WD2 corresponding to the first risk warning WD1 is obtained), vehicle C1 further identifies the received object detection result OD (S340). The subsequent operations are similar to those in the first example and will not be elaborated here.

[0100] Based on the above description, it can be known that the method for identifying false risk warnings in a vehicle workshop network and the in-vehicle electronic device using the method provided by the present invention have the following advantages: (1) Low computational load: The method for identifying false risk warnings in a vehicle workshop network provided by the present invention does not require additional arithmetic operations at all times. By using the data provided by the advanced driver assistance system, it can be determined whether the first risk warning is a fake risk warning.

[0101] (2) Low latency and low power consumption: Since there is no need to perform additional arithmetic operations at all times, the latency and power consumption of the in-vehicle electronic device 100 can also be reduced.

[0102] (3) Higher reliability: Since the method for identifying false risk warnings in a vehicle workshop network provided by the present invention uses the existing data provided by the original advanced driver assistance system 200, it avoids the reliability problems caused by specification matching.

[0103] Based on the above, the method for identifying false risk warnings in a vehicle workshop network and the in-vehicle electronic device using the method provided by the present invention can determine whether the first risk warning is trustworthy based on object detection operation, the first risk warning based on the vehicle workshop network, and the second risk warning based on object detection operation, and then identify the false first risk warning, avoiding wrong driving behaviors due to corresponding false first risk warnings and ensuring the safety of the vehicle. In addition, since it is only when the first risk warning based on the vehicle workshop network is obtained that it is verified whether the first risk warning is trustworthy, there is no need to continuously compare objects in all surrounding environments at all times to avoid risk warnings caused by false vehicle workshop data, reducing the resource consumption of the in-vehicle electronic device and thus improving driving efficiency.

Claims

1. A method for identifying false risk warnings in a workshop network, applicable to in-vehicle electronic devices of vehicles, wherein the in-vehicle electronic device includes a processor and a communication circuit unit, and the in-vehicle electronic device is connected to the workshop network via the communication circuit unit. The method includes: Instructing the advanced driver assistance system of the vehicle to perform object detection operations; In response to obtaining a first risk warning based on the workshop network, determining whether a second risk warning based on the object detection operation corresponding to the first risk warning is obtained, wherein the second risk warning is received from the advanced driver assistance system; In response to determining that the second risk warning based on the object detection operation corresponding to the first risk warning is obtained, determining that the first risk warning based on the workshop network is trustworthy; In response to determining that the second risk warning based on the object detection operation corresponding to the first risk warning is not obtained, determining whether the first risk warning based on the workshop network is trustworthy according to the object detection results of the object detection operation and multiple preset conditions corresponding to the object detection operation; In response to determining that the first risk warning based on the workshop network is trustworthy, adjusting the driving behavior of the vehicle according to the first risk warning or the second risk warning; And In response to determining that the first risk warning based on the workshop network is not trustworthy, ignoring the first risk warning and not adjusting the driving behavior of the vehicle according to the first risk warning.

2. The method for identifying false risk warnings in a workshop network according to claim 1, wherein the types of the first risk warning and the second risk warning include: Forward Collision Warning (FCW); And Blind Spot Warning (BSW).

3. The method for identifying false risk warnings in a workshop network according to claim 2, wherein the step of determining whether a second risk warning based on the object detection operation corresponding to the first risk warning is obtained includes: Determining whether the type of the first risk warning is the same as the type of the received second risk warning; In response to determining that the type of the first risk warning is different from the type of the received second risk warning, determining that the second risk warning based on the object detection operation corresponding to the first risk warning is not obtained; In response to determining that the type of the first risk warning is the same as the type of the received second risk warning, determining whether the risk object of the first risk warning corresponds to the risk object of the second risk warning; In response to determining that the risk object of the first risk warning corresponds to the risk object of the second risk warning, determining that the second risk warning based on the object detection operation corresponding to the first risk warning is obtained; And In response to determining that the risk object of the first risk warning does not correspond to the risk object of the second risk warning, determining that the second risk warning based on the object detection operation corresponding to the first risk warning is not obtained.

4. The method for identifying false risk warnings in a workshop network according to claim 1, wherein the object detection operation includes one or more of the following operations: First object detection operation based on image data; Second object detection operation based on point cloud data: and Third object detection operation based on mixed data, where the mixed data includes the image data and the point cloud data.

5. The method for identifying false risk warnings in a workshop network according to claim 4, wherein the step of determining whether the first risk warning based on the workshop network is trustworthy according to the object detection result of the object detection operation and the plurality of preset conditions corresponding to the object detection operation includes: When the object detection operation is the first object detection operation based on the image data, identify the first object detection result of the first object detection operation and the first object feature threshold value and the first object probability threshold value corresponding to the first object detection operation, wherein the first object detection result includes the detected first object and the first object feature and the first object probability corresponding to the detected first object, and the first object feature threshold value and the first object probability threshold value are the plurality of first preset conditions corresponding to the first object detection operation; Determine whether the first object feature is greater than the first object feature threshold value; Determine whether the first object probability is greater than the first object probability threshold value; In response to determining that the first object feature is greater than the first object feature threshold value and the first object probability is greater than the first object probability threshold value, determine that the first risk warning based on the workshop network is trustworthy; And In response to determining that the first object feature is not greater than the first object feature threshold value or the first object probability is not greater than the first object probability threshold value, determine that the first risk warning based on the workshop network is untrustworthy.

6. The method for identifying false risk warnings in a workshop network according to claim 5, wherein the first object feature includes the image feature size or the image feature ratio of the detected first object, and the image feature ratio of the detected first object is the ratio of the image feature size of the detected first object to the preset image size of the image data, and the first object probability is the existence probability of the detected first object.

7. The method for identifying false risk warnings in a workshop network according to claim 4, wherein the step of determining whether the first risk warning based on the workshop network is trustworthy according to the object detection result of the object detection operation and the plurality of preset conditions corresponding to the object detection operation includes: When the object detection operation is the second object detection operation based on the point cloud data, identify the second object detection result of the second object detection operation and the second object distance threshold value and the second object probability threshold value corresponding to the second object detection operation, wherein the second object detection result includes the detected second object and the second object distance and the second object probability corresponding to the detected second object, and the second object distance threshold value and the second object probability threshold value are the plurality of second preset conditions corresponding to the second object detection operation; Determine whether the second object distance is less than the second object distance threshold value; Determine whether the second object probability is greater than the second object probability threshold value; Responsive to determining that the distance of the second object is less than the second object distance threshold value and the probability of the second object is greater than the second object probability threshold value, it is determined that the first risk warning based on the in-vehicle network is trustworthy; And Responsive to determining that the distance of the second object is not less than the second object distance threshold value or the probability of the second object is not greater than the second object probability threshold value, it is determined that the first risk warning based on the in-vehicle network is untrustworthy.

8. The method for identifying a false risk warning in an in-vehicle network according to claim 7, wherein the distance of the second object is the distance between the detected second object and the vehicle, and the probability of the second object is the probability of the existence of the detected second object.

9. The method for identifying a false risk warning in an in-vehicle network according to claim 4, wherein the step of determining whether the first risk warning based on the in-vehicle network is trustworthy according to the object detection result of the object detection operation and the plurality of preset conditions corresponding to the object detection operation includes: When the object detection operation is the third object detection operation based on the mixed data, identify the third object detection result of the third object detection operation and the third object feature threshold value, the third object distance threshold value, and the third object probability threshold value corresponding to the third object detection operation, wherein each third object detection result includes a detected third object and the third object feature, the third object distance, and the third object probability corresponding to the detected third object, and the third object feature threshold value, the third object distance threshold value, and the third object probability threshold value are a plurality of third preset conditions corresponding to the third object detection operation; Determine whether the third object probability is greater than the third object probability threshold value, wherein responsive to determining that the third object probability is not greater than the third object probability threshold value, it is determined that the first risk warning based on the in-vehicle network is untrustworthy, wherein responsive to determining that the third object probability is greater than the third object probability threshold value, determine whether the third object feature is greater than the third object feature threshold value, wherein responsive to determining that the third object feature is greater than the third object feature threshold value, it is determined that the first risk warning based on the in-vehicle network is trustworthy, wherein responsive to determining that the third object feature is not greater than the third object feature threshold value, determine whether the third object distance is less than the third object distance threshold value, wherein responsive to determining that the third object distance is less than the third object distance threshold value, it is determined that the first risk warning based on the in-vehicle network is trustworthy, wherein responsive to determining that the third object distance is not less than the third object distance threshold value, it is determined that the first risk warning based on the in-vehicle network is untrustworthy.

10. The method for identifying a false risk warning in an in-vehicle network according to claim 9, wherein the third object feature includes the size or ratio of the image feature of the detected third object, and the ratio of the image feature of the detected third object is the ratio of the size of the image feature of the detected third object to the preset image size of the image data, and the third object probability is the probability of the existence of the detected third object.

11. An in-vehicle electronic device, applicable to a vehicle, comprising: A communication circuit unit for connecting to a vehicle-to-vehicle network; A processor coupled to the communication circuit unit; And A storage circuit unit storing instructions which, when executed by the processor, configure the in-vehicle electronic device to perform the following actions: Instruct the advanced driver assistance system of the vehicle to perform object detection operations; In response to obtaining a first risk warning based on the vehicle-to-vehicle network, determine whether a second risk warning based on the object detection operations corresponding to the first risk warning is obtained, wherein the second risk warning is received from the advanced driver assistance system; In response to determining that the second risk warning based on the object detection operations corresponding to the first risk warning is obtained, determine that the first risk warning based on the vehicle-to-vehicle network is trustworthy; In response to determining that the second risk warning based on the object detection operations corresponding to the first risk warning is not obtained, determine whether the first risk warning based on the vehicle-to-vehicle network is trustworthy according to the object detection results of the object detection operations and a plurality of preset conditions corresponding to the object detection operations; In response to determining that the first risk warning based on the vehicle-to-vehicle network is trustworthy, adjust the driving behavior of the vehicle according to the first risk warning or the second risk warning; And In response to determining that the first risk warning based on the vehicle-to-vehicle network is not trustworthy, ignore the first risk warning and do not adjust the driving behavior of the vehicle according to the first risk warning.

12. The in-vehicle electronic device according to claim 11, wherein the types of the first risk warning and the second risk warning include: Forward Collision Warning (FCW); And Blind Spot Warning (BSW).

13. The in-vehicle electronic device according to claim 12, wherein the step of determining whether the second risk warning based on the object detection operations corresponding to the first risk warning is obtained includes: Determine whether the type of the first risk warning is the same as the type of the received second risk warning; In response to determining that the type of the first risk warning is different from the type of the received second risk warning, determine that the second risk warning based on the object detection operations corresponding to the first risk warning is not obtained; In response to determining that the type of the first risk warning is the same as the type of the received second risk warning, determine whether the risk object of the first risk warning corresponds to the risk object of the second risk warning; In response to determining that the risk object of the first risk warning corresponds to the risk object of the second risk warning, determine that the second risk warning based on the object detection operations corresponding to the first risk warning is obtained; And In response to determining that the risk object of the first risk warning does not correspond to the risk object of the second risk warning, determine that the second risk warning based on the object detection operations corresponding to the first risk warning is not obtained.

14. The in-vehicle electronic device according to claim 11, wherein the object detection operations include one or more of the following operations: The first object detection operation based on image data; The second object detection operation based on point cloud data: and The third object detection operation based on mixed data, where the mixed data includes the image data and the point cloud data.

15. The in-vehicle electronic device according to claim 14, wherein the step of determining whether the first risk warning based on the vehicle-to-vehicle network is trustworthy according to the object detection result of the object detection operation and the plurality of preset conditions corresponding to the object detection operation includes: When the object detection operation is the first object detection operation based on image data, identifying the first object detection result of the first object detection operation and the first object feature threshold value and the first object probability threshold value corresponding to the first object detection operation, wherein the first object detection result includes the detected first object and the first object feature and the first object probability corresponding to the detected first object, and the first object feature threshold value and the first object probability threshold value are a plurality of first preset conditions corresponding to the first object detection operation; Determining whether the first object feature is greater than the first object feature threshold value; Determining whether the first object probability is greater than the first object probability threshold value; In response to determining that the first object feature is greater than the first object feature threshold value and the first object probability is greater than the first object probability threshold value, determining that the first risk warning based on the vehicle-to-vehicle network is trustworthy; And In response to determining that the first object feature is not greater than the first object feature threshold value or the first object probability is not greater than the first object probability threshold value, determining that the first risk warning based on the vehicle-to-vehicle network is not trustworthy.

16. The in-vehicle electronic device according to claim 15, wherein the first object feature includes the image feature size or the image feature ratio of the detected first object, and the image feature ratio of the detected first object is the ratio of the image feature size of the detected first object to the preset image size of the image data, and the first object probability is the existence probability of the detected first object.

17. The in-vehicle electronic device according to claim 14, wherein the step of determining whether the first risk warning based on the vehicle-to-vehicle network is trustworthy according to the object detection result of the object detection operation and the plurality of preset conditions corresponding to the object detection operation includes: When the object detection operation is the second object detection operation based on point cloud data, identifying the second object detection result of the second object detection operation and the second object distance threshold value and the second object probability threshold value corresponding to the second object detection operation, wherein the second object detection result includes the detected second object and the second object distance and the second object probability corresponding to the detected second object, and the second object distance threshold value and the second object probability threshold value are a plurality of second preset conditions corresponding to the second object detection operation; Determining whether the second object distance is less than the second object distance threshold value; Determining whether the second object probability is greater than the second object probability threshold value; Responsive to determining that the distance of the second object is less than the second object distance threshold and the probability of the second object is greater than the second object probability threshold, it is determined that the first risk warning based on the in-vehicle network is trustworthy; and Responsive to determining that the distance of the second object is not less than the second object distance threshold or the probability of the second object is not greater than the second object probability threshold, it is determined that the first risk warning based on the in-vehicle network is untrustworthy.

18. The in-vehicle electronic device according to claim 17, wherein the distance of the second object is the distance between the detected second object and the vehicle, and the probability of the second object is the probability of the existence of the detected second object.

19. The in-vehicle electronic device according to claim 14, wherein the step of determining whether the first risk warning based on the in-vehicle network is trustworthy according to the object detection result of the object detection operation and the plurality of preset conditions corresponding to the object detection operation includes: When the object detection operation is the third object detection operation based on the mixed data, identifying the third object detection result of the third object detection operation and the third object feature threshold, the third object distance threshold, and the third object probability threshold corresponding to the third object detection operation, wherein the third object detection result includes the detected third object and the third object feature, the third object distance, and the third object probability corresponding to the detected third object, and the third object feature threshold, the third object distance threshold, and the third object probability threshold are a plurality of third preset conditions corresponding to the third object detection operation; judging whether the third object probability is greater than the third object probability threshold, wherein responsive to determining that the third object probability is not greater than the third object probability threshold, it is determined that the first risk warning based on the in-vehicle network is untrustworthy, wherein responsive to determining that the third object probability is greater than the third object probability threshold, judging whether the third object feature is greater than the third object feature threshold, wherein responsive to determining that the third object feature is greater than the third object feature threshold, it is determined that the first risk warning based on the in-vehicle network is trustworthy, wherein responsive to determining that the third object feature is not greater than the third object feature threshold, judging whether the third object distance is less than the third object distance threshold, wherein responsive to determining that the third object distance is less than the third object distance threshold, it is determined that the first risk warning based on the in-vehicle network is trustworthy, wherein responsive to determining that the third object distance is not less than the third object distance threshold, it is determined that the first risk warning based on the in-vehicle network is untrustworthy.

20. The in-vehicle electronic device according to claim 19, wherein the third object feature includes the size or ratio of the image feature of the detected third object, the ratio of the image feature of the detected third object is the ratio of the size of the image feature of the detected third object to the preset image size of the image data, and the third object probability is the probability of the existence of the detected third object.

Images